Premium
Discussion of ‘Estimates of extreme values by different distribution functions’ by K. C. Majumdar and R. P. Sawhney [Comment]
Author(s) -
Weiss L. L.
Publication year - 1966
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/wr002i001p00165-01
Subject(s) - mathematics , extreme value theory , distribution (mathematics) , prime (order theory) , statistics , mathematical analysis , combinatorics
Majumdar and Sawhney [1965] present very useful comparisons of some distribution functions widely used in hydrological practice. The comparison made by expressing the ratio of the exceedance probabilities is most practical. Because the extreme‐value and the log‐normal distributions have perhaps the largest number of devotees, the ratios of equation 12 presented in Figure 1 are of prime interest. Following equation 12 the authors state that the ratio depends ‘only’ on the coefficient of variation. This relationship holds true, of course, only if K T is calculated from equation 10 or 11, which implicitly assume that the period of record ( n ) is infinite. When that is not the case, the ratio is dependent on n as well as C v